...

The HL7 FHIR standard has many benefits over some previous HL7 approaches and is gaining a lot of attention and implementation. There is also FHIR-hype, usually not from the core team behind FHIR, but from others hoping that FHIR will solve (almost) all healthcare information interoperability needs. We will highlight some differences and commonalities between the FHIR and openEHR approaches and exemplify how context of use and political/business views influences the short- and long-term benefits of different options and combinations. Some systems based on openEHR are succesfully used in IHE profiled exchange environments and some are IHE-certified, we'll also discuss such options and combinations.

2.2.2 Working with openEHR Semantically (Koray Atalag)

We have used openEHR to model and persist experimental data that underpins computational physiology models (e.g. VPH, Human Physiome). The idea is then to link both experimental and real-world clinical information to these quantitative and predictive models to create a new breed of decision support tools that can deliver highly personalised and precision medicine. We have had some key important learnings while representing such models and especially when semantically annotating them - which in openEHR world corresponds to term and constraint bindings and data instance level term mappings. We will explain our methodology and discuss lessons learned which we hope will facilitate the use of openEHR in Semantic Web environments.

2.2.3 EtherCIS a new Open Source OpenEHR backend server (Christian Chevalley)

EtherCIS is a DB centric openEHR compliant SOA platform. It exposes a RESTful API handling OpenEhr entities under three formats (Flat json, AQLPath json and XML). The motivation of EtherCIS has been to provide an open, scalable yet secure clinical data repository that can be queried using different languages (AQL and SQL). SQL querying is performed using standard tools, as recent progresses in DB  developments now allow combinations of relational and schema-less data structures. We will discuss our experiences (and failures) with previous persistence approaches, pros and cons of the current one and future perspectives in line with major DB engines developments.

2.2.4 3 Implementation of an openEHR repository using a Graph Database (Samar El Helou)

Building openEHR repositories is challenging since it requires a thorough grasp and implementation of the openEHR Reference Model (RM) that has numerous classes in a tree-structure with deep hierarchy. Moreover, a mismatch between the database model and the RM can lead to high development time and cost. The graph model shares many semantic similarities with the definitions of the openEHR RM making it a potential fit for its representation and implementation. We will propose a method for implementing an openEHR repository by a graph database employing the labeled property graph model. We will also discuss some limitations and opportunities of persisting openEHR data with Neo4j.

2.2.5 4 Clinical Knowledge Manager (Sebastian Garde)  

To be able to exchange clinical information in a semantically safe way across different openEHR-based systems, it is important to agree on the clinical concepts used in these systems. In openEHR, such concepts are formally expressed in archetypes and developed in regional, national and international collaboration. It is crucial that clinicians - even without any knowledge of openEHR - are inherently involved in this process by being able to review and comment as required. Only this can ensure that the clinical content models are clinically valid and comprehensive. To enable this collaboration, the Clinical Knowledge Manager (CKM) has been developed as a web-based system for collaborative development, management, validation, review and publishing of openEHR archetypes and other clinical knowledge resources. CKM is used internationally by the openEHR foundation as well as in several regional and national programmes. CKM supports the 'federation' of archetypes, so that the various programmes can work independently and to their own timelines, while sharing archetypes with each other where possible.

3. Workshop speakers

• Shinji Kobayashi, MD, PhD - Kyoto University, Japan
Koray Atalag (Deactivated), MD, PhD, FACHI - University of Auckland, New Zealand
• Sebastian Garde, Dr. sc. hum., Dipl.-Inform. Med., FACHI - Ocean Informatics, Germany
• Erik Sundvall, MSc, PhD - Linköping University and Region Östergötland, Sweden
• Christian Chevalley, Founder and Technical Director, ADOC Software, Thailand
• Samar El Helou - Kyoto University, Japan
• Borut Fabjan, Marand, Slovenia

...